LocationsDisaster risk management
Disaster risk management
Extreme weather events such as floods, earthquakes and fires are increasing in frequency and they compound other crises like food and water insecurity. The impacts from extreme weather events hit the poorest countries hardest as these are particularly vulnerable to damage caused by environmental hazards and may need more time to rebuild and recover. Between 2000 and 2019 over 475,000 people lost their lives as a direct result of extreme weather events globally. Floods are one of the most common climate-related hazards and an estimated 1.47 billion people globally are exposed to flooding’s substantial risks. But extreme heat is a growing threat, particularly in cities, and current climate trajectories will mean at least twice as many megacities could become heat stressed, exposing more than 350 million people to deadly heat by 2050. Mitigating and responding to climate-related disasters poses a number of challenges due to their complexity and the fact that the needs of crisis-affected communities outstrip available humanitarian resources.
Collective intelligence initiatives are attempting to tackle this problem in two main ways. The first is by addressing several data gaps. Early warning systems and enhanced emergency preparedness and response are critical to reducing loss and damage from climate disasters, but both require good quality, localized data to ensure relevant and effective targeting of resources. Crowdsourcing is creating localized data on hazards, which, in combination with other sources including official data or sensor data, helps improve the precision of local forecasts, early warning systems and risk models. In some examples, collective intelligence goes beyond addressing the data gap to target doing gaps – enabling both frontline responders and affected communities to take more coordinated and effective action during a crisis. They do this by sharing hazard or vulnerability data with communities to raise awareness and build capacity, or by providing enhanced situational awareness and early warning alerts to officials through digital dashboards.
Main collective intelligence methods being used
Combining citizen-generated data with official data or sensor data
Crowdsourcing data and collaborative modeling to improve scientific models of flood risks
Main climate action gaps being addressed
- Data gap around real-time, localized data about climate-related disasters
- Doing gap from poor coordination and ineffective targeting of resources during disaster response
Citizen-generated data (in combination with other data) for early warning, preparedness and response to disasters
The UN Secretary General’s Early Warnings for All initiative has focused international attention on the importance of investing in early warning systems, which can help to dramatically reduce large financial losses from climate disasters. But to generate the forecasts needed to ensure everyone is protected by early warning systems by 2027 it is necessary to address several gaps in weather observation.
Collective intelligence methods such as crowdmapping and crowdsourcing are currently filling some of these data gaps on climate hazards, infrastructure and weather – expanding the range of available data sources that can be used to provide early warnings and situational awareness of a crisis. An example of this is PhilAWARE, a hazard monitoring and early warning system to improve disaster management and decision making in the Philippines. It was built using local infrastructure data mapped with local and global volunteers through Humanitarian OpenStreetMap. It consolidates hazard information and alerts from various sources and disseminates alerts to officials and impacted communities to help them take action.
Similarly, in Uruguay, the Monitor Integral de Riesgos y Afectaciones (MIRA), is an integrated disaster management and early warning system that combines several official and citizen-generated datasets. It gathers social media data and crowdsourced reports about the impacts of disasters on homes, goods and services for improved situational awareness, and issues text alerts directly to affected communities.
Community Water Watch is a community designed and operated early warning service focusing on flooding in Dar es Salaam, where floods are a constant threat and often result in fatalities. The service collects information about real-time flooding by web-scraping online news and crowdsourcing reports from affected people via a chatbot on the messaging app Telegram. It combines these data with hydrometeorological data collected through low-cost sensors to provide situational reports about the precise location of floods. These are shared with frontline responders at the Tanzanian Red Cross and their volunteers so they can respond more quickly and target their support to where it is most needed.
The Living Lab of West Africa takes a different approach to flood management in Ouagadougou, Burkina Faso, a city that's been affected by increasingly frequent and intense flash floods. The project used crowdmapping to identify waste dumping sites that were blocking drains and installed low-cost rain gauges to collect water level data in strategic locations across the city. Combining these datasets and official data, they plan to build a forecasting model to provide early warnings to responders and local residents. The initiative has brought together residents, government officials and municipal services to collaborate on more effective responses to flooding. The Lab also organizes skills and capacity building sessions where residents learn new approaches to land rehabilitation and composting, and participate in waste cleanup to reduce the impact of flash floods. These activities raise awareness about the links between personal behaviors, climate change and local impacts.
Crowdsourcing data and collaborative modeling to improve scientific models of flood risks
Collective intelligence initiatives are also creating novel datasets about flooding – unprecedented in scale and granularity – through crowdsourcing of videos and localized knowledge. This improves the ability of scientists to develop accurate, high resolution flooding models to understand the risks posed by flash floods.
Floodchasers is a project that crowdsources videos of flash floods into a centralized database for hydrology researchers. It was created due to the insufficiency of information on flash floods in urban rivers and basins. Videos are submitted by members of the public and responders at the frontlines of flooding events. This dataset is helping scientists better understand river behaviors and flooding patterns, and improving the calibration of water flow models that are used to create forecasts and warnings.
Another example is an initiative run by Deltares in Tanzania with the Tanzania Red Cross Society, the World Resources Institute and others. They work with local residents to map and label infrastructure in flood-prone areas of Dar Es Salaam using OpenStreetMap. Residents draw on their local knowledge to add detail about buildings and their elevation, creating a high quality hazard dataset of the area. Researchers have used this dataset and causal information gathered during collaborative modeling workshops with participants to develop a flood risk model with street level precision, a higher resolution than is normally possible.
